Re: [PATCH] Free swap suspend from dependency on PageReserved

["Rafael J. Wysocki" <rjw@sisk.pl>]

Hi Nigel,

On Thursday, 6 of October 2005 00:24, Nigel Cunningham wrote:
> Hi Dave.
> 
> On Thu, 2005-10-06 at 08:03, Dave Hansen wrote:
> > On Thu, 2005-10-06 at 07:04 +1000, Nigel Cunningham wrote:
> > > 
> > > +       for (tmp = 0; tmp < max_low_pfn; tmp++, addr += PAGE_SIZE) {
> > > +               if (page_is_ram(tmp)) {
> > > +                       /*
> > > +                        * Only count reserved RAM pages
> > > +                        */
> > > +                       if (PageReserved(mem_map+tmp))
> > > +                               reservedpages++;
> > 
> > Please don't reference mem_map[] directly outside of #ifdef
> > CONFIG_FLATMEM areas.  It is not defined for all config cases.  Please
> > use pfn_to_page(), instead.  It should work in all cases where the page
> > is valid.
> > 
> > Also, instead of keeping addr defined like you do, and comparing it
> > during each run of the loop, why not just use pfn_is_nosave(), which is
> > already defined?  Then, you won't even need the variable.
> 
> Thanks for the comments. Will revise and resend.

There's one more reference to PageReserved() in swsusp.c, which is
in save_highmem_zone().  You seem to omit this one (or I have missed
it in your patch in which case please ignore this message).  Is that
intentional?

Rafael